18 BULLETIN 1339, TJ. S. DEPARTMENT OF AGRICULTURE 



MIDDAY DECLINE IN RATE OF GAIN 



During the period of spring honey flow, after the morning loss in 

 weight has ceased, which may occur at any time between 6.30 and 

 10 a. m. (the average hour for 1922 was 8 a. m., and for 1923 9 a. m.), 

 the colony increases in weight with a more or less regular accelera- 

 tion until midday. At this time a decided slackening usually occurs 

 in the rapidity of gain in weight. This change in rate of gain is sur- 

 prisingly constant in the time of its appearance and can be distin- 

 guished during practically the whole of the spring honey flow. The 

 minimum rate of gain in weight, after gains have been established 

 in the morning, usually comes at about 1 or 2 o'clock in the after- 

 noon. On some days this decrease in rate of gain is decided, while 

 on other days it is scarcely noticeable. When actual changes in 

 weights are plotted this midday decrease in rapidity of gain can 

 scarcely be seen, but when the data are plotted so as to show differ- 

 ences in weight from hour to hour it becomes quite obvious (figs. 

 3,7,8). 



The cause of this change in rate of gain at midday is not entirely 

 clear. Dufour (10) in his records mentions a similar phenomenon, 

 and Bonnier (1, p. 163) connects a decrease in the amount of available 

 nectar with low relative humidity and high temperatures prevailing at 

 the same time of day. Bonnier measured the amount of nectar pro- 

 duced at various hours and found that less nectar is produced toward 

 the middle of the day, when high temperatures and low relative humid- 

 ities are usually encountered. He further found that at this time 

 bees return to the hive with less than their maximum loads, and that 

 fewer bees leave the hive for the field during this midday period, 

 in comparison with periods before and after. Either of these two 

 facts might be sufficient to account for the midday decline in rate of 

 gain. 



Figure 5, as stated, shows the averftge hourly changes in weight for 

 colonies 1, 2, and AB for the spring honey flow. The similarity in the 

 graphs for colonies 1 and 2 is not surprising, since these colonies were 

 side by side and each was stimulated by exactly the same outside 

 factors. The graph for colony AB, however, represents the changes 

 for a colony one year earlier, but it closely follows the others. The 

 midday decline in rate of gain is prominent in all three, and, although 

 differing in magnitude, the three are intimately similar from hour to 

 hour. The graphs for average temperature, relative humidity, and 

 hours of sunshine bear no visible relation to the break in the three 

 graphs showing gain and loss. The maximum temperatures for both 

 years come later in the day than the midday decline, and the minimum 

 relative humidity occurs at an hour slightly preceding. Until further 

 investigations are made, Bonnier's (1, p. 164) observation of decreased 

 nectar secretion on the part of the plant and the resulting effect on 

 bee behavior must be accepted as the most logical explanation of 

 this phenomenon, although the reasons which he gives for these 

 changes on the part of the plants and the bees do not seem entirely 

 satisfactory. 



NET GAIN 



The midday decline in rate of gain is included as part of the net 

 gain, since an actual loss rarely occurs at this time during a good 

 honey flow. The 13 graphs in Figure 6, for as many days, show 



